Howdy 👋

In my 2017 book Artificial Intelligence for .NET (co-authored with Nishith), I described how language models of that era worked. A combination of gradient descent and supervised fine-tuning gave them predictive and classification superpowers for the given domain/context. Those models were next token predictors in the truest sense.

The attention/transformers architecture together with data-science-optimized GPUs revolutionized language models and made today’s LLMs possible. The fundamentals of pre-training and teaching models have stayed the same.

It’s incredible how a single breakthrough in software architecture can lead to generational advancements.

Beautifully written article by Alex Xu, so easy to digest. Highly recommended even if you aren’t familiar with the technical details.

TIL that there’s a name for the inter-service communication pattern that I implemented in our microservices solution five years ago.

It’s called Choreography-based Saga. This is where each service listens for interesting events from other services to trigger actions.

The alternative is Orchestration-based Saga where a central orchestrator listens to events from and issues commands to services.

It works! This is how I approached my first Jira ticket at my new job. Rather than waiting to digest the entire codebase or for code walkthroughs to happen, I simply dove into unfamiliar waters and created a pull request on day one. I suddenly had specific, pointed questions to the team to inform my approach rather than generic questions that might not have been helpful in solving the problem.